id_set.hpp

Go to the documentation of this file.

#ifndef OSMIUM_INDEX_ID_SET_HPP
#define OSMIUM_INDEX_ID_SET_HPP

/*

This file is part of Osmium (http://osmcode.org/libosmium).

Copyright 2013-2017 Jochen Topf <jochen@topf.org> and others (see README).

Boost Software License - Version 1.0 - August 17th, 2003

Permission is hereby granted, free of charge, to any person or organization
obtaining a copy of the software and accompanying documentation covered by
this license (the "Software") to use, reproduce, display, distribute,
execute, and transmit the Software, and to prepare derivative works of the
Software, and to permit third-parties to whom the Software is furnished to
do so, all subject to the following:

The copyright notices in the Software and this entire statement, including
the above license grant, this restriction and the following disclaimer,
must be included in all copies of the Software, in whole or in part, and
all derivative works of the Software, unless such copies or derivative
works are solely in the form of machine-executable object code generated by
a source language processor.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE, TITLE AND NON-INFRINGEMENT. IN NO EVENT
SHALL THE COPYRIGHT HOLDERS OR ANYONE DISTRIBUTING THE SOFTWARE BE LIABLE
FOR ANY DAMAGES OR OTHER LIABILITY, WHETHER IN CONTRACT, TORT OR OTHERWISE,
ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
DEALINGS IN THE SOFTWARE.

*/

#include <algorithm>
#include <cassert>
#include <cstddef>
#include <cstring>
#include <iterator>
#include <memory>
#include <type_traits>
#include <vector>

#include <osmium/osm/item_type.hpp>
#include <osmium/osm/types.hpp>

namespace osmium {

    namespace index {

        template <typename T>
        class IdSet {

        public:

            virtual ~IdSet() = default;

            virtual void set(T id) = 0;

            virtual bool get(T id) const noexcept = 0;

            virtual bool empty() const = 0;

            virtual void clear() = 0;

            virtual std::size_t used_memory() const noexcept = 0;

        }; // class IdSet

        template <typename T>
        class IdSetDense;

        template <typename T>
        class IdSetDenseIterator {


            const IdSetDense<T>* m_set;
            T m_value;
            T m_last;

            void next() noexcept {
                while (m_value != m_last && !m_set->get(m_value)) {
                    const T cid = IdSetDense<T>::chunk_id(m_value);
                    assert(cid < m_set->m_data.size());
                    if (!m_set->m_data[cid]) {
                        m_value = (cid + 1) << (IdSetDense<T>::chunk_bits + 3);
                    } else {
                        const auto slot = m_set->m_data[cid][IdSetDense<T>::offset(m_value)];
                        if (slot == 0) {
                            m_value += 8;
                            m_value &= ~0x7;
                        } else {
                            ++m_value;
                        }
                    }
                }
            }

        public:

            using iterator_category = std::forward_iterator_tag;
            using value_type        = T;
            using pointer           = value_type*;
            using reference         = value_type&;

            IdSetDenseIterator(const IdSetDense<T>* set, T value, T last) noexcept :
                m_set(set),
                m_value(value),
                m_last(last) {
                next();
            }

            IdSetDenseIterator<T>& operator++() noexcept {
                if (m_value != m_last) {
                    ++m_value;
                    next();
                }
                return *this;
            }

            IdSetDenseIterator<T> operator++(int) noexcept {
                IdSetDenseIterator<T> tmp{*this};
                operator++();
                return tmp;
            }

            bool operator==(const IdSetDenseIterator<T>& rhs) const noexcept {
                return m_set == rhs.m_set && m_value == rhs.m_value;
            }

            bool operator!=(const IdSetDenseIterator<T>& rhs) const noexcept {
                return ! (*this == rhs);
            }

            T operator*() const noexcept {
                assert(m_value < m_last);
                return m_value;
            }

        }; // class IdSetDenseIterator

        template <typename T>
        class IdSetDense : public IdSet<T> {


            friend class IdSetDenseIterator<T>;

            // This value is a compromise. For node Ids it could be bigger
            // which would mean less (but larger) memory allocations. For
            // relations Ids it could be smaller, because they would all fit
            // into a smaller allocation.
            constexpr static const std::size_t chunk_bits = 22;
            constexpr static const std::size_t chunk_size = 1 << chunk_bits;

            std::vector<std::unique_ptr<unsigned char[]>> m_data;
            T m_size = 0;

            static std::size_t chunk_id(T id) noexcept {
                return id >> (chunk_bits + 3);
            }

            static std::size_t offset(T id) noexcept {
                return (id >> 3) & ((1 << chunk_bits) - 1);
            }

            static unsigned char bitmask(T id) noexcept {
                return 1 << (id & 0x7);
            }

            T last() const noexcept {
                return static_cast<T>(m_data.size()) * chunk_size * 8;
            }

            unsigned char& get_element(T id) {
                const auto cid = chunk_id(id);
                if (cid >= m_data.size()) {
                    m_data.resize(cid + 1);
                }

                auto& chunk = m_data[cid];
                if (!chunk) {
                    chunk.reset(new unsigned char[chunk_size]);
                    ::memset(chunk.get(), 0, chunk_size);
                }

                return chunk[offset(id)];
            }

        public:

            using const_iterator = IdSetDenseIterator<T>;

            IdSetDense() = default;

            bool check_and_set(T id) {
                auto& element = get_element(id);

                if ((element & bitmask(id)) == 0) {
                    element |= bitmask(id);
                    ++m_size;
                    return true;
                }

                return false;
            }

            void set(T id) final {
                (void)check_and_set(id);
            }

            void unset(T id) {
                auto& element = get_element(id);

                if ((element & bitmask(id)) != 0) {
                    element &= ~bitmask(id);
                    --m_size;
                }
            }

            bool get(T id) const noexcept final {
                if (chunk_id(id) >= m_data.size()) {
                    return false;
                }
                auto* r = m_data[chunk_id(id)].get();
                if (!r) {
                    return false;
                }
                return (r[offset(id)] & bitmask(id)) != 0;
            }

            bool empty() const noexcept final {
                return m_size == 0;
            }

            T size() const noexcept {
                return m_size;
            }

            void clear() final {
                m_data.clear();
                m_size = 0;
            }

            std::size_t used_memory() const noexcept final {
                return m_data.size() * chunk_size;
            }

            IdSetDenseIterator<T> begin() const {
                return {this, 0, last()};
            }

            IdSetDenseIterator<T> end() const {
                return {this, last(), last()};
            }

        }; // class IdSetDense

        template <typename T>
        class IdSetSmall : public IdSet<T> {

            std::vector<T> m_data;

        public:

            void set(T id) final {
                m_data.push_back(id);
            }

            bool get(T id) const noexcept final {
                const auto it = std::find(m_data.cbegin(), m_data.cend(), id);
                return it != m_data.cend();
            }

            bool get_binary_search(T id) const noexcept {
                return std::binary_search(m_data.cbegin(), m_data.cend(), id);
            }

            bool empty() const noexcept final {
                return m_data.empty();
            }

            void clear() final {
                m_data.clear();
            }

            void sort_unique() {
                std::sort(m_data.begin(), m_data.end());
                const auto last = std::unique(m_data.begin(), m_data.end());
                m_data.erase(last, m_data.end());

            }

            std::size_t size() const noexcept {
                return m_data.size();
            }

            std::size_t used_memory() const noexcept final {
                return m_data.capacity() * sizeof(T);
            }

            using const_iterator = typename std::vector<T>::const_iterator;

            const_iterator begin() const noexcept {
                return m_data.cbegin();
            }

            const_iterator end() const noexcept {
                return m_data.cend();
            }

            const_iterator cbegin() const noexcept {
                return m_data.cbegin();
            }

            const_iterator cend() const noexcept {
                return m_data.cend();
            }

        }; // class IdSetSmall

        template <template<typename> class IdSetType>
        class NWRIdSet {

            using id_set_type = IdSetType<osmium::unsigned_object_id_type>;

            id_set_type m_sets[3];

        public:

            id_set_type& operator()(osmium::item_type type) noexcept {
                return m_sets[osmium::item_type_to_nwr_index(type)];
            }

            const id_set_type& operator()(osmium::item_type type) const noexcept {
                return m_sets[osmium::item_type_to_nwr_index(type)];
            }

        }; // class NWRIdSet

    } // namespace index

} // namespace osmium

#endif // OSMIUM_INDEX_ID_SET_HPP